Work

Llama-3.1-FoundationAI-SecurityLLM-Base-8B Technical Report

Year: 2025

Type: preprint

Abstract: As transformer-based large language models (LLMs) increasingly permeate society, they have revolutionized domains such as software engineering, creative writing, and digital arts. However, their adoption in cybersecurity remains limited due to challenges like scarcity of specialized training data and complexity of representing cybersecurity-specific knowledge. To address these gaps, we present Foundation-Sec-8B, a cybersecurity-focused LLM built on the Llama 3.1 architecture and enhanced through continued pretraining on a carefully curated cybersecurity corpus. We evaluate Foundation-Sec-8B across both established and new cybersecurity benchmarks, showing that it matches Llama 3.1-70B and GPT-4o-mini in certain cybersecurity-specific tasks. By releasing our model to the public, we aim to accelerate progress and adoption of AI-driven tools in both public and private cybersecurity contexts.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2504.21039

Authors: Kassianik, Paul, Saglam, Baturay, Chen, Alexander, Nelson, Blaine, Vellore, Anu, Aufiero, Massimo, Burch, Fraser, Kedia, Dhruv, Zohary, Avi, Weerawardhena, Sajana, Priyanshu, Aman, Swanda, Adam, Chang, Amy, Anderson, Hyrum, Oshiba, Kojin, Santos, Omar, Singer, Yaron, Karbasi, Amin

Language: English

Topic: Adversarial Robustness in Machine Learning

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W4417521649

Cites:

Cited by:

Related to: